Market researchers may often be considered ‘mad’ by our colleagues in other departments, but when I speak of bipolar and unipolar I am not referring to a psychiatric disorder. Instead my thinking is on scales, those oft employed measurements of attitudes and perceptions that we as survey authors create to uncover what our customers, prospects, and employees are thinking.
Scales can be unipolar or bipolar. Unipolar scales refer to the presence or absence of an attribute. They are typically shorter and less taxing on the respondent than bipolar scales. This makes them an ideal choice for surveys, particularly for ones that may have issues with cognitive load. Customer satisfaction can be measured using a single pole of reference. For example,
Bipolar scales rely on two opposite poles. They require the respondent to balance these opposite attributes. Typically they are coded using seven points (1 to 7 or -3 to 0 to +3). The middle point is reserved for an absence of either quality (neither dissatisfied nor satisfied for example.) Satisfaction can be measured in a bipolar method by assessing the presence of dissatisfaction or satisfaction with a neutral point in the middle. Another example would be a semantic differential as these scales require the use of bipolar adjectives to measure feelings toward an object or person.
Both scales have their plusses and minuses. Unipolar scales tend to be easier on the respondent, whereas bipolar scales offer potentially greater variation. Testing both scale options is an effective method for seeing which variation works best with your respondents.
Written by Greg Timpany